» Lire en français: French
Diane Pierret and Roberto Steri share a keen research interest in the regulatory environment for banking as is evident from their recent co-authored paper Stressed Banks. In this Q&A, they talk about the paper, the post-crash regulatory environment for banking, and some potentially serious implications of the proposed Financial CHOICE Act in America, both for risk taking in US banking and the stability of the global financial system.
5 min read
Am I right in thinking that, following the financial crisis that began in 2007-2008, the Dodd-Frank Act in the US introduced some important innovations to America’s banking regulations?
Diane: Prior to the Act banks were already subject to capital requirements, although to a lesser extent. However, the Act introduced a new measure of testing the strength of banks.
The Dodd–Frank Wall Street Reform and Consumer Protection Act (Dodd-Frank) introduced an annual stress test for certain banks conducted by the Federal Reserve (The Fed). A number of large banks are subject to more invasive screening of their balance sheet and more stringent capital requirements based on the sensitivity of their assets to a stress scenario.
Not all banks are stress tested?
Roberto: Not all the banks are subject to the regulatory stress test performed by the Fed, although other banks do perform stress tests themselves. The stress testing by the Fed is restricted to a number of banks that have more than $50bn of assets.
Broadly speaking how does that stress testing regime work?
Diane: The banks are required to submit granular data on their whole balance sheet, and in particular on their asset holdings. The Fed defines a stress scenario and sets a required capital ratio (threshold). Then the Fed will project the stress scenario on the granular bank data it has collected. The result will be a lower capital ratio under the stress scenario since capital has to absorb the projected losses under the stress scenario. If a bank has a capital ratio under stress that falls below the threshold then it is subject to sanctions. Constraints are imposed on the bank’s capital plans, such that it might not be allowed to distribute dividends anymore.
The Fed implements the stress tests (i.e., uses its own model to derive projected losses under stress), which is not the case with most central banks in other jurisdictions. We believe, as do many people in the financial services industry, that this is one of the main innovations in Dodd-Frank.
However as part of the proposed Financial CHOICE Act in the US the Fed’s stress testing regime is due to be rolled back?
Roberto: Yes, we are now at a point where the Senate in the US is due to vote on the Financial CHOICE Act. And although there is little chance that the Act will pass in full, it is possible that while capital requirements may become more stringent, at the same time banks will be able to be exempted from stress testing.
And your research has investigated some possible implications of rolling back the stress testing regime, while increasing capital requirements?
Diane: Essentially, in our paper we look at the stressed banks, those banks that are subject to a tougher regulatory environment than the non-stressed banks, and investigate how these banks react to the stress-testing regime.
Roberto: During the crisis banks were taking too much risk. They were lending to risky companies, for example.
We investigate the impact of the introduction of more stringent regulations, which were designed to create a safer system that discourages risky lending, and where banks will refrain from investing in risky assets. However, as our research reveals, the impact of such measures, the effect on bank behaviour, is not simple. A number of factors need to be considered in order to understand what the real impact of the regulations are on the type of investment and risk that banks engage in after stress testing is introduced.
What kinds of factors?
Diane: Well even though the regulator wants to make the system safe they need to take into account how banks respond to the regulation when they are, at the same time, having in mind the need to achieve their goal of making money – of maximising profits.
So in your paper you distinguish between two aspects of the regulatory environment – stringent capital requirements and macro prudential stress testing. You look at their impact both separately and when working in tandem. And consider whether they lead, overall, to riskier or safer banking behaviour. What do you find?
Roberto: At its most basic banks collect money and invest money to make a profit. If you place restrictions on them that increase the cost of collecting that money, if you force them to issue equity and adjust their capital ratio, then you make it more difficult for them to make profits through non-risky investments.
Making money in the financial markets is not easy. That implies that the banks, in order to stay profitable, must take on more risk. And this is largely what we find in the data. Banks that are subject to tighter capital requirements tend to take on more risk.
So potentially the very measures designed to discourage risky behaviour by the banks actually has the opposite effect?
Roberto: Exactly. In the free market the consequences of regulatory measures are different than intended.
It is important to note that we are not saying that stress testing is ineffective, though. Overall if you are subject to stress tests and you are monitored, you tend to limit your risky behaviour. We see, for example, that the risk of the stressed banks since Dodd-Frank has been reduced more than that of the non-stressed banks because the banks that are punished more are those that make risky investments.
However it is also true that, in order to survive, even stressed banks need to be profitable. And those banks are trying to compensate for the increase in the cost of funding, the cost of capital requirements – the costs that come from stress tests – that is probably why they dislike stress tests so much!
Then your findings suggest that, by increasing capital requirements and rolling back stress testing, the Financial CHOICE Act might make matters worse?
Diane: Our paper shows that the main guard against banks taking on additional risk because of a high capital requirement is the stress test. If you remove that you are basically providing a strong incentive to make risky investments while giving banks the freedom to make those investments. That cannot be contributing to greater financial instability.
So what would be your message to the people in power, to the Senators, the policymakers?
Diane: Before the financial crisis there were no stress tests, but there were capital requirements, although probably not as stringent. But the financial crisis happened anyway. Why? Because the investments that banks made were not monitored properly; the Fed had much less power to regulate and monitor the banks.
Dodd-Frank has enabled the Fed economists to go on site (at the banks) and review the risks that banks are exposed to, examine their risk models, and interview risk management teams. Yes it takes resources but there is a big advantage in terms of having a more informed view on the riskiness of banks’ investments.
The current system may be imperfect – we don’t know everything that a bank is doing. Banks will still seek ways to take risks that are not observable by the Fed. But we are definitely in a better position than before the financial crisis.
Roberto: Imperfect monitoring is still better than no monitoring at all.
Diane: But if the measures in the Financial CHOICE Act go through then it is a step backwards. Worse, because there will be even higher capital requirements and no monitoring, there will be even greater incentives to take risks.
Yes you should maximise the ratio of how much information is gathered through the stress tests to the resources involved in conducting the stress tests, but at the very least you need to monitor the riskiness of large banks’ investments if you want to keep the system safe.
And although directly relevant to the US, given the impending Financial CHOICE Act, are your findings applicable outside the US?
Diane: We looked at the US because the data is available to study the banks that undergo stress testing there, however our findings can be generalised to any banking system.
So your research might be helping to make the global financial system a safer place and avoid a repeat of the financial crisis?
Roberto: Well we certainly want to raise awareness of the issues involved so policy makers can take better informed decisions
Related research paper: D. Pierret, R. Steri, Stressed Banks (2017)